Why Implementation of Predictive Maintenance can be Challenging
Companies that use IoT gather large volumes of data but most of the time are not prepared enough to tap into its full potential, which is a waste. In service organizations, this valuable data can be used for predictive maintenance. A survey by PwC held in 2018 showed that only 11% of companies can be considered mature in the field of preventative maintenance. This was the same level as in the previous survey that was held in 2017!
Predictive maintenance not only ensures more efficient business processes but unmistakably leads to higher customer satisfaction and thus retention! Still, it seems that a lot is being said and written about the Internet of Things (IoT) and Big Data, but too often it is left at that.
One definition of operational excellence is to fulfill your customer’s needs to the best of your ability. To be successful at that, it is paramount to first carefully listen to and think about the needs of your potential customer. Does the customer want air conditioning or a pleasant indoor climate? If the customer wants a pleasant indoor climate, this means that you will need to offer a total solution instead of a single product.
By equipping a heating boiler with smart sensors, it can become possible to predict when maintenance will be needed through the approach of predictive maintenance. Data collected by smart sensors can be compiled and analyzed automatically and then used to generate accurate maintenance predictions.
An excellent example of the use of an implementation of predictive maintenance is German elevator manufacturer ThyssenKrupp. The company uses sensors to measure factors like engine temperature, the speed of the elevator cab, the performance of the door and so on. Analysis of all this data allows the company to perform the necessary maintenance before the elevator shuts down, preventing quite a few claustrophobic moments!
The new gold
If IoT is the enabler and Big Data is what flows from that, how to start? To tap into the new gold of Big Data for the purpose of predictive maintenance, it is crucial to determine thresholds. Thresholds are values that will trigger an action when they are exceeded. When these thresholds are exceeded, automatic choices can be made consequently, by for example using Machine Learning. If for instance an escalator generates data on the number of rotations, the speed, the total weight of people and the air humidity, and there is a combination of low humidity and high weight, the system can automatically disperse more oil onto the rotation axels, or perhaps create a work order to schedule a service technician to do preventative maintenance.
A 2019 Survey by ReliablePlant shows that only a little more than 5% of companies use cloud-based platforms for their predictive maintenance. To make your life easier, it is essential to pool collected data into one single platform in the cloud. Service companies are starting to see the benefits of one central data hub. In addition to the fact that data is processed quickly and more efficiently using a cloud approach, advanced technology is also more available and easily adapted to specific purposes from the cloud. System upgrades and updates are always available to cloud users first. To be up-to-date and have access to your data from any location, working from the cloud is a huge asset.
One of our customers that have already made significant steps in the field of predictive maintenance is Royal Brinkman, a global specialist in horticulture. Royal Brinkman desired to have control over the use of fertilizer liquids and to optimize logistical processes. The implemented IoT project enabled them to collect data about weather conditions and predictions, actual humidity, temperature, etc. This data is stored at a central hub and compared across different sources, after which conclusions are made. That way Royal Brinkman knows the exact use of fertilizer liquids at any time and is able to predict use in the near future. Will the humidity be lower next week? Then more fertilizer liquid needs to be added and a tank will be empty sooner. Using this knowledge, Royal Brinkman employees do not need to refill tanks every day, but they can see exactly when a tank will run dry. This has resulted in the optimization of logistical processes and purchasing, thus reducing costs.
Previously, ‘service’ meant solving the customer’s instant problems. By now, IoT and Big Data have made it possible to change the meaning of the concept of service by ‘solving the customer’s problems before they have occurred’. Predictive maintenance is key to optimizing internal processes, boosting service levels and ultimately increasing customer satisfaction. Microsoft Dynamics 365 Finance and Supply Chain Management with a dedicated solution on top, might just be the enabler for your company’s ability to fulfill that tomorrow’s customers already know what they want today.